God and First Person in Berkeley

Transcription

1 1 God and First Person in Berkeley by George Botterill Note: This is a draft version of a paper whose final and definitive form is as published in: Philosophy 82 nr.319 (January 2007), pp For online access to the published paper, go to: The state of our understanding of how Berkeley s argument for the existence of God is related to his idealism constitutes something of a puzzle. His argument for God s existence plays a significant and explicit role in Berkeley s philosophical system a feature in which his main philosophical works are quite dissimilar to Locke s, for example. So Berkeley s argument for the existence of God ought to be of central interest in his philosophy, and the quality of the argument should be important for assessment of Berkeley s standing as a philosopher. However, there is no clear agreement as to how exactly his argument is to be reconstructed, and it is quite possible to doubt whether any of the familiar reconstructions get Berkeley s reasoning right. In the end I believe this is a puzzle for which there can be no ordinary solution, because there is a deep reason why Berkeley s argument resists attempts at fully satisfactory reconstruction. That reason is a sort of equivocation or shift in perspective in Berkeley s use of the first person. And that equivocation or perspective-shift (I hope this characterisation will be satisfactorily clarified in what follows) runs through the whole of Berkeley s case for idealism, and indeed without it the pretensions of Berkeleian immaterialism and idealism 1 to be a bulwark against scepticism would be exposed as somewhat flimsy. That is the awkward and difficult part of the case to be presented. However, it is not the place to start. We must rather start by trying to find an argument for God s existence which would have seemed powerful and convincing to Berkeley. I suggest we can indeed find a compelling argument to the existence of God from Berkeleian idealism, an argument with which it is quite understandable that Berkeley should have been pleased. For those sympathetic to Berkeley this is the bright side of the story. But there is also a darker aspect. For if his reasoning had been made fully explicit, it would have exposed the fault-line in his philosophical project 2. For the moment it will be expedient to proceed as if we are merely trying to set out the best argument available to Berkeley. In order to do this we need to improve upon the various interpretations already on offer. It is not entirely clear how we should enumerate these interpretations. There are two standard labels, the Continuity Argument and the Passivity Argument 3, for possible interpretations of how Berkeley sought to prove God s existence, the differentiation of these two arguments being largely due to an influential article by Jonathan Bennett. 4 Jonathan Dancy also discerns two possible lines of argument in Berkeley s texts, though he also suggests that these two arguments are not really separable in the way they might at first sight seem to be. 5 Dancy refers to the two interpretations as the independence argument and the continuity argument, but it is not clear that his independence argument is

2 2 significantly different from Bennett s passivity argument. Stoneham offers an Independence Argument which is distinguishable, and intended to be distinguished, from the Passivity Argument, since it appeals to considerations of ontological dependence rather than a premise about the causation of ideas. 6 Yet for all their various differences, these several interpretations seem to suffer from much the same basic defect, and a severe defect at that. They fall short as interpretations of something Berkeley might have regarded as at once a novel argument and a positive advantage to his idealism, for one simple reason. This is that, even if we allow Berkeley all their premises, they fail to establish the existence of a single, unique spirit with the divine attributes. This paper will attempt to offer what appears to be a more satisfactory interpretation of Berkeley s argument, an interpretation which, in the spirit of a principle of charity, may at first seem considerably more creditable to Berkeley s philosophical reputation. However, it also serves to highlight certain limitations to his overall philosophical project, which are otherwise liable to escape attention. Exactly how a principle of charity should in general inform interpretation of philosophical texts is a matter of some dispute. One might suspect that the charity of philosophical interpreters is a selective form of generosity: they endorse the reasoning of their target writer in one area only to set up some refutation or objection elsewhere. I would have supposed that the application of a principle of charity to Berkeley s argument for the existence of God was relatively unproblematic. But others may disagree. So I will state the parameters within which the inquiry in this paper will proceed. They are as follows. We are not under an obligation to credit him with a sound argument for the existence of God, that would be demanding too much. But we must not represent him as merely reworking a traditional argument. This is because his argument for God should both depend upon his idealism and should also be a novel argument. Since we do not (I shall presume) endorse idealism, we will not share all Berkeley s premises and so will not be inclined to accept his argument. The result we are aiming for is an argument of which we might say: Since I am not an idealist (I do not think that objects are collections of ideas), I need not accept the conclusion. But if I were convinced of the truth of idealism, then I would recognise the strength of the argument. Two points of clarification need to be added to these parameters. The first is that there is no requirement that the argument we assign to Berkeley should be deductively valid. On the contrary, I think it is crucial that we see Berkeley s argument for the existence of God as a case of inference to the best explanation. There seems to be a willingness on the part of several commentators to read Berkeley as offering an inference to the best explanation. But of course it is also important that we identify both the correct target phenomenon (or explanandum) and the correct conclusion. 7 The second is that these parameters assume that the case for idealism is to be argued on grounds independent of the attempted proof of God s existence: that the latter depends upon the former, whereas the former does not depend upon the latter. This is a view of Berkeley s philosophical system which some might wish to contest. They might maintain that Berkeley s philosophy is theocentric not just in the sense that the argument for God s existence is of central interest to Berkeley, but that it is equally at the centre of his argument for idealism. Against this I can only say here that this is not how Berkeley presents matters. In both the Principles of Human

3 3 Knowledge 8 and the Three Dialogues the arguments for idealism are advanced first and he subsequently follows up with his argument for the existence of God. It is true that he thereby strengthens the overall explanatory coherence of his position. But he is arguing from the claim that real things (sensible objects) are dependent for their existence upon minds. In summary, there are three requirements which a satisfactory interpretation of Berkeley s argument for the existence of God should meet. Firstly, it should be different from any of the traditional arguments for the existence of God which would already have been familiar to Berkeley and his readers (thus justifying his claim to be advancing a new argument). Secondly, it should be a powerful argument, given that Berkeleian idealism is accepted. And, thirdly, it should explain the freedom with which Berkeley moves between use of the first person singular and the first person plural in presenting his argument. Meeting this third requirement provides the key to what is distinctive about Berkeley s argument. In what follows I start by outlining other interpretations and explaining why I think they are unsatisfactory. I then propose what I take to be a superior interpretation of Berkeley s view, which we may term the Shared Reality Argument. I attempt to substantiate this interpretation by appeal both to overall considerations of argumentative strategy and to details of Berkeley s texts. The complication here is that the adequacy of the interpretation is affected by the way in which we take Berkeley to be using the first person. I argue that his use of the first person is not genuinely Cartesian. I conclude by considering what entitlement Berkeley has to believe in the phenomena for which his argument for the existence of God might provide the best explanation. I. The Arguments: From Continuity, Passivity, and Independence to Shared Reality In the Second Dialogue Berkeley has Philonous pointing out to Hylas that it is an advantage of idealism that it furnishes a direct and immediate demonstration, from a most evident principle, of the being of a God. 9 But what is that most evident principle? The sentence before puts the idealist argument for the existence of God in this pithy form: Sensible things do really exist; and, if they really exist, they are necessarily perceived by an infinite Mind: therefore there is an infinite Mind or God. (Three Dialogues between Hylas and Philonous, Second Dialogue, ) This is a summary of his argument which commentators have found rather challenging to unpack. It is commonly suggested that the once standard interpretation of Berkeley s argument associated the real existence of sensible objects with their continued existence, during the intervals between any finite spirits having ideas of them. This, of course, is the interpretation so memorably satirised in Ronald Knox s limerick about the tree in the quad. This line of reasoning, known as the Continuity Argument, can be formulated as follows:

4 4 The Continuity Argument 1. Neither a single idea nor any collection of ideas can exist unless perceived by some spirit. 2. Objects are nothing other than collections of ideas. 3. Objects do continue to exist even when they are not being perceived by any human spirit. Therefore, there is a non-human spirit who perceives objects when human spirits are not doing so. This interpretation of Berkeley may come naturally to students seeking to find a way to reconcile Berkeleian idealism with commonsense convictions about the continued endurance of objects. But it is not at all easy to identify passages in Berkeley s texts which lend themselves to being read in a straightforward way as presentations of the Continuity Argument. The best candidate for such a presentation would appear to be an exchange from the Third Dialogue in which Hylas asks Philonous if he can conceive it possible that sensible objects should exist if he were to be annihilated. The reply is: I can; but then it must be in another mind. When I deny sensible things an existence out of the mind, I do not mean my mind in particular, but all minds. Now it is plain they have an existence exterior to my mind, since I find them by experience to be independent of it. There is therefore some other mind wherein they exist, during the intervals between the times of my perceiving them: as likewise they did before my birth, and would do after my supposed annihilation. (Three Dialogues between Hylas and Philonous, Third Dialogue, 230-1) Since Berkeley had already presented versions of his novel argument for the existence of God in several earlier passages (PHK paragraphs and paragraph 146; and also at some length in the Second Dialogue), this constitutes, as Bennett pointed out, only tenuous evidence for attributing the Continuity Argument to Berkeley. It is also doubtful whether this argument really deserves its recognition as the once traditional interpretation, at least among scholarly commentators on Berkeley. For example, Luce and Mabbott 11 did not subscribe to it. Whether or not the Continuity Argument was the standard interpretation before Jonathan Bennett s 1965 paper, after that article it has come to be discussed as one major alternative to the argument which Bennett considers to be Berkeley s main argument, and which he formulates as follows: The Passivity Argument 1. Ideas of sense come into my mind without being caused by any act of my will. 2. The occurrence of any idea must be caused by the will of some being in whose mind the idea occurs. Therefore, my ideas of sense are in the mind of, and caused by the will of, some being other than myself. 12 The Continuity and Passivity Arguments are clearly distinct. They are, indeed, so very different that it would be less than charitable to suggest Berkeley might have confounded them. It remains possible, of course, that he might have taken the Passivity Argument to be his main argument, and then added the Continuity

5 5 Argument as further reinforcement. This is far from an attractive understanding of Berkeley s intentions, however, since both these arguments should appear deeply problematic to any critical reader. The Continuity Argument is the more obviously defective of the two. Some will point out that the conclusion that God perceives objects when finite spirits are not doing so cannot be acceptable, given that Berkeley holds perception to be the passive reception of ideas. But it is not clear how weighty this point is, since despite the esse est percipi slogan Berkeley is more clearly committed to the existence of objects being mind-dependent than that the dependence should always be realised in the sort of perceptions finite spirits enjoy: divine perception might be quite different. (And consider the way in which he takes conceiving of an object to establish its being in a mind and so not truly unperceived in the Master Argument, as also in entry 472 in the Philosophical Commentaries.) But there is no doubt that attention has rightly been focussed upon the third premise of the Continuity Argument. This premise invites the challenge: Why is Berkeley entitled to help himself to the commonsense assumption that objects continue to existence in the intervals between human perceptions? That is indeed something everyone believes. But, without matter, what is the case for continuity? The Passivity Argument does seem closer to Berkeley s texts and is less glaringly defective. Even so, its second premise does prompt such questions as: How can Berkeley establish that ideas cannot come into existence except through the volitional activity of spirits? Or if he thinks this is something he has already proved, where exactly is that proof to be found? These are questions of considerable interest in their own right. But we should not allow their interest to distract us from the significant point that, as formulated above, the Continuity and Passivity Arguments share a common defect. And it is a fatal defect, at that: they do not justify the conclusion that God exists. It is not as if commentators fail to notice this. Bafflingly, they notice it but treat it almost as if it were a side issue in the evaluation of the argumentation. One way of underestimating the defect is to allow that the God whose existence Berkeley is engaged in proving is not clearly to be identified with the God of Christianity. Thus it may be queried that while Berkeley s God is a metaphysical support and sustainer of the whole order of existence, is he therefore to be regarded as perfectly benevolent and worthy of worship? This issue is remarked but dismissed in a startlingly offhand way by Bennett with the comment: Nor shall I consider the yawning gulf between the conclusion of the argument and the Christian monotheism which it is supposed to serve. This gulf, and the moves which Berkeley might make to bridge it, are matters of routine apologetics which have little philosophical interest. 13 Grayling takes the matter more seriously. His verdict is: Berkeley s desire to identify the metaphysical God required by his system with the traditional God required by his religious commitments is simply a result of those religious commitments, and is otherwise unsubstantiated. 14 But whether the argument can yield a conclusion about the existence of a God with the attributes of traditional theology is a secondary matter in relation to the main gulf between the premises and the conclusion of these two arguments. The main failing of these two arguments is not that the spirit who perceives when finite spirits are not

6 6 doing so, or the spirit who produces ideas of sense experienced by finite spirits, might fail to match the conception of what the Christian God is. It is rather that the arguments just fail to deliver the conclusion that there is such a spirit. This is sometimes demurely put by saying that they fail to establish uniqueness. But this is a disastrous shortcoming. Only if he had committed the sort of elementary fallacy involved in inferring that there is someone who is universally loved from the premise that everybody loves somebody, could Berkeley have imagined that such an argument as either the Passivity or Continuity Argument proved the existence of God. Without such a switch in quantifier-scope (i.e., from assigning widest scope to the universal quantifier to assigning widest scope to the existential quantifier), there is no way in which Berkeley or anybody else could arrive at the conclusion that there is a single non-human spirit which perceives or produces all these ideas (ideas either not perceived or not produced by human minds). A weakness common to both Continuity and Passivity Arguments will not help to pick out either of those two as the preferred interpretation of Berkeley s reasoning. But this point might well be of some importance in spurring us to consider some other argument that is free from this weakness. By distinguishing ontological dependence from causal dependence, Stoneham is able to discern two arguments for the existence of God in Berkeley s Three Dialogues. One of these is just the Passivity Argument, though, as noted above, Stoneham refers to it as the causal argument. 15 The other is the Independence Argument. He derives his original formulation of this argument from the following passage: To me it is evident, for the reasons you allow of, that sensible things cannot exist otherwise than in a mind or spirit. Whence I conclude, not that they have no real existence, but that seeing they depend not on my thought, and have an existence distinct from being perceived by me, there must be some other mind wherein they exist. As sure therefore as the sensible world really exists, so sure is there an infinite, omnipresent Spirit who contains and supports it. (Three Dialogues between Hylas and Philonous, Second Dialogue, 212) Stoneham suggests that this argument can be summed up as follows: The Independence Argument in experience we are aware of certain items (ideas) which are ontological dependants (not being self-creating), but are not dependent upon me, therefore there exists something else upon which they depend. Since we are talking about ideas, which are mind-dependent, the only thing they could depend on would be a mind. 16 But even without going into any detailed consideration of how the Independence Argument differs from the Passivity Argument, we can see at once that this is not going to help with the main difficulty. The Independence Argument, as presented by Stoneham, has exactly the same fatal defect as the Continuity and Passivity Arguments. Stoneham more or less concedes this point, commenting on the alleged reasoning: we should question whether Berkeley has established uniqueness, let alone the other theological qualities. 17 But this way of proceeding is less than charitable. Unless we are forced to do so, we should not read Berkeley as offering a futile piece of argumentation, much less spend time debating which of several futile and clearly fallacious lines of reasoning it is better to assign to Berkeley.

7 7 Ayers seizes on the very same passage as Stoneham, and in particular the claim that sensible things... have an existence distinct from being perceived by me, as a source for something he calls the Distinctness Argument. 18 Depending upon how exactly distinctness is to be understood, this may be a step in the direction of the interpretation advocated here. But since Ayers is primarily concerned with attacking Bennett s interpretation, by showing both that Bennett was wrong in thinking that the Passivity Argument was Berkeley s main argument and also wrong in restricting existence distinct from being perceived by me to the sort of continued existence that the Continuity Argument assumes, it is not entirely clear how big a step Ayers is taking. Having reviewed the proffered interpretations of Berkeley s arguments, the only verdict we can return is that they would expose Berkeley as failing completely in his purpose of advancing a novel argument for the existence of God. It is, indeed, only by using sleight of hand in the concealment of a blatant fallacy that they even succeed in giving the impression that he so much as has an argument for God s existence, rather than an argument for an unspecified number of non-human spirits. If this were the best that could be done for one of Berkeley s most cherished arguments, then his standing in the historical philosophical canon would be brought into question. But we can do better. It is time to introduce an argument which really does deliver the conclusion Berkeley seeks. So consider instead an argument which I will refer to as the Shared Reality Argument : The Shared Reality Argument 1. Our ideas of the sensible world are not caused by any act of our wills (because perception is passive). 2. The occurrence of any idea must be caused by the will of some spirit. 3. Our ideas are intricately co-ordinated in a harmonious and reliable way, so as to enable us to perceive a common reality. Therefore, our ideas of sense are caused by the will of an incomprehensibly great and omniscient spirit. In some respects this argument seems quite close to the Passivity Argument. It shares the assumption that any idea must be brought into being by the volition of some spirit. It differs in being expressed in the first person plural rather than the first person singular. This, however, is a crucially important difference. A spirit which engendered ideas only in Berkeley s mind would not be so easy to distinguish from Descartes s Demon. 19 But Berkeley s God has the awe-inspiring power to sustain a massive and perfect integration of the ideas of sense in the minds of all his creatures. The argument is also to be taken as an inference to the best explanation. Those of us who are not idealists will have no difficulty over rejecting it, because we reject the second premise of the argument. If we believe in material objects, then we have a general explanation for the intricate co-ordination of human sensory experience: it is the result of our perceiving the same physical items, which causally impact upon our senses in predictable ways. But if we had rejected the very existence of mindindependent physical objects, the Shared Reality Argument would surely seem compelling. Compare it, for example, with arguments which attempt to establish that

8 8 we should recognise a single author of both the Iliad and the Odyssey on grounds of style, composition and thematic affinity. Such arguments can make the hypothesis of a single author fairly probable, but no more than that. But if the whole of nature as presented to human experience is produced by spiritual agency, then the conclusion that it is the work of a single spirit of supreme power and knowledge, is beyond reasonable doubt. In terms of the three conditions outlined above, the Shared Reality Argument fits the bill. It certainly seems to be a novel argument. For while it is true that it could just possibly be regarded as a variant form of the Argument from Design (the grand design being common access to a real order of nature), it is a variant never encountered before Berkeley. Nor is it an argument which needs to be supplemented by a traditional argument, at least in so far as the conclusion of the existence of an allknowing and all-powerful God is concerned. For the point of the argument is that the integration of ideas of sense in the minds of human spirits and other finite creatures requires complete information about what all these beings have experienced and are experiencing, combined with an ability to produce at will exactly appropriate new ideas of sense in their minds. So if this were Berkeley s argument, then he does indeed have an original argument and the first requirement is met. The second requirement was that it should be a strong argument, given that Berkeley is to be allowed the truth of idealism as a premise. And the Shared Reality Argument is a strong argument from idealism to the existence of God, even though it is likely to leave those unconvinced by Berkeley s arguments against matter unmoved. That is just the result we should be seeking. And of course it does not suffer from the main defect of the Continuity, Passivity, and Independence Arguments, which is certainly desirable. Galileo and Descartes may have maintained the Book of Nature was written in the language of mathematics. Berkeley was particularly concerned to prove it had a single author. The third requirement concerned the way Berkeley formulates his argument, sometimes using the first person singular and sometimes using the first person plural. This is a feature of Berkeley s texts which stands in need of explanation, but is usually passed over in silence by commentators, who themselves adopt either singular or plural formulations with no clear justification for preferring one or the other. For example, the Continuity Argument is regularly formulated (as above) in the plural whereas Bennett presented the Passivity argument in the first person singular. It might seem clear that the Shared Reality Argument cannot be adequately expressed in the first person singular. For how can premises concerning my own ideas of sense do justice to the third premise of the Shared Reality Argument, which asserts that our ideas are suitably correspondent and co-ordinated, thus allowing us to perceive a single real world? I will return to this important point in the next section, in which I hope to explain why Berkeley moves as freely as he does between first person singular and first person plural. Whatever its merits as an argument from idealism to the existence of God, the Shared Reality Argument may strike readers of Berkeley as a surprising, implausible, and even extravagant interpretation of his argument. It must be admitted that its crucial third premise does not sound very much like anything Berkeley ever wrote or would ever have written. I make no apologies for this, since I have deliberately phrased the

9 9 argument in a way intended to make clear what force it has for someone who accepts idealism, rather than using terms calculated to echo Berkeley s own texts. So in response to this point it is possible to retreat to the claim that, even if nothing at all like this argument had ever occurred to Berkeley, it would still be of philosophical interest to formulate the way in which he should have reasoned to God s existence from his idealism. However, is it necessary to retreat to that weaker claim? The objection that this does not sound much like Berkeley (or that we have no direct textual evidence in favour of this interpretation) is less worrying than it might at first seem, for a number of reasons. The first is that Berkeley does have a way of talking about the integration and co-ordination of ideas of sense. He calls these phenomena the laws of nature, as in this passage: the set rules or established methods, wherein the mind we depend on excites in us the ideas of sense, are called the laws of nature (PHK 30). We need to remind ourselves that Berkeley s conception of the laws of nature is really quite distinctive and is wholly phenomenological. From the perspective of modern science the explanation of regularities in our perceptual experiences is going to be an extremely complicated matter for secondary sciences, depending upon the functioning of all sorts of causal mechanisms in objects perceived, intervening media, and the physiology and psychology of the perceivers. But for Berkeley these are Nature s basic regularities, and their connection with God s existence should be apparent to us: It is therefore plain, that nothing can be more evident to any one that is capable of the least reflexion, than the existence of God, or a spirit who is intimately present to our minds, producing in them all that variety of ideas or sensations, which continually affect us, on whom we have an absolute and entire dependence, in short, in whom we live, and move, and have our being. (PHK 149) He reiterates this conception of massive phenomenological regularity and coordination in the next paragraph, responding to the suggestion that Nature might produce these ideas in our minds by questioning whether we can mean anything by Nature except the visible series of effects, or sensations imprinted on our minds according to certain fixed and general laws (PHK 150). Even when he is not explicit about the lawlike regularities manifested in ideas of sense, everything he takes to be real must involve co-ordination of ideas of sense in different minds, since they have an existence distinct from being perceived by me : real objects outstrip any collection of ideas within a single finite mind. This is what Berkeley means when he says such things as: The ideas imprinted on the senses by the Author of Nature are called real things (PHK 33); and then goes on to describe them as orderly and coherent. A second reason is that while inference to the best explanation does require that the explanatory target be discerned and indicated, it is not always necessary or even possible to specify all the features which warrant the conclusion of such reasoning. A wound, for example, may have features which, to a trained eye, make it very clear how it must have been inflicted. But it may not be so easy to describe exactly what these features are. And it certainly is not easy to specify what the co-ordination of ideas of sense is which makes them the perception of a common reality. Any attempt to give a more detailed description of this massive co-ordination is liable to seem hopelessly inadequate. A further reason why the absence of explicit formulation of the third premise of this argument in Berkeley s texts should not count against this interpretation is that in one

10 10 of the texts the presentation of the argument in the form of a dialogue provides a context in which the co-ordination of ideas of sense in different minds is naturally presupposed. When Hylas is about to concede to scepticism, Philonous draws his attention to what the world as we experience it is like. But of course he can only do this if it is assumed that they are given a common object of attention through the coordination of their ideas. The false imaginary glare passage of the Second Dialogue can appear something of an embarrassment, as if Berkeley were trying to embellish his argument for God s existence by waxing lyrical about the wonders of His creation. Certainly he is moved to describe the whole system as immense, beautiful, glorious beyond expression and beyond thought!. But, by his own doctrine, in referring to the world as we know it he is referring to ideas in the minds of finite creatures, and he has Philonous make claims about what ideas they have, for example, of stars: Yet if you take the telescope, it brings into your sight a new host of stars that escape the naked eye. Here they seem contiguous and minute, but to a nearer view immense orbs of light at various distances, far sunk in the abyss of space. (Three Dialogues between Hylas and Philonous, Second Dialogue, 211) So there are strong reasons to suppose that even if Berkeley did not explicitly formulate his argument in such a way as to make it clear that he was employing the Shared Reality Argument, this piece of reasoning was definitely a part of his novel, immaterialist case for the existence of God. As far as the strength of the argument is concerned, once one begins to appreciate the phenomena to be explained, the Shared Reality Argument is seen to be about as compelling an instance of inference to the best explanation as could be found. It is a very strong argument, even if not demonstrative. However, in criticising Atherton s claim that Berkeley s argument is an inference to the best explanation 20 Stoneham objects that the inclusion of the premise that ideas can only be caused by spirits prevents this from being such an inference, since that premise precludes any alternative explanation. He insists that: There is a big difference between a probabilistic argument from effects to causes and a demonstrative argument with an a priori premiss about the nature of causation. 21 Now this objection is partly motivated by Stoneham s belief that Berkeley s idealism is itself offered as the conclusion of an inference to the best explanation, or at any rate an inference to a better explanation than that our ideas of sense are caused by material objects. It must be doubtful whether this line of argument can give Berkeley a strong enough version of immaterialism, since it is important to him that he should be able to claim that matter is not merely a superfluous hypothesis, but something that could not possibly exist. 22 But leaving that issue to one side, it is clear this is not a demonstrative argument because the Shared Reality Argument really is intended to establish uniqueness. Its conclusion is not that our ideas of sense are produced by a non-human spirit or spirits, but that our ideas of sense are produced by a single spirit who knows all our minds. Nor is this argument intended to establish idealism, though no doubt in Berkeley s eyes it served to reinforce it. For Berkeley believed he had already shown that matter could not produce ideas of sense. And it is entirely legitimate in an inference to the best explanation to rely upon what else you take yourself to know in drawing a conclusion. For example, if I see what looks like my signature on some document, but know that I never signed it, then it may well be reasonable for me to infer that my signature has been forged.

11 11 There is, however, an obvious difficulty with the Shared Reality Argument. What is Berkeley s entitlement to its third premise? It is by no means apparent why he should be on safer ground here in assuming that there are a multitude of other finite spirits who have ideas of a shared reality which are systematically and intricately related with Berkeley s own (and, what s more, the ideas of each one inter-related with everyone else s) than in assuming the continuity of sensible objects. Ultimately Berkeley has to be convicted of assuming more than he is entitled to, by his own lights, on this point. One might describe this as a sort of circularity in his reasoning, even though that involves a certain smoothing of the tangled knot produced by shifting from first person singular to first person plural. The tangle can even be seen in his modern interpreters, as shown in the following curious comment by Winkler, which echoes what is going on in Berkeley s own text rather than explicating it: Yet I do not think it follows that Berkeley s argument for God s existence can be commenced only after the existence of other finite minds has been established. It would be better to say that the existence of God and other finite minds should be inferred simultaneously, as the best total explanation of our ideas of sense. 23 Of course, if the Shared Reality Argument is the correct interpretation of Berkeley s reasoning, a simultaneous inference to God and other finite minds is quite inadmissible. But what can Winkler mean by the best total explanation of our ideas of sense? Perhaps he just means that each and every one of us can, if he (or she) reflects intently enough, see that the existence of God and of other finite minds is the best total explanation of his (or her) own ideas of sense. That is not a very natural way of thinking about our ideas of sense. Rather than each person being restricted to appealing to his or her own ideas of sense, we have a much better argument for God s existence if any one of us can infer to an explanation of all our ideas of sense. But that, of course, is a first person plural perspective, not a position from which the existence of other finite minds can be inferred in a non-question-begging way. It is at this point that philosophical charity turns nasty. We have conceded Berkeley a strong argument. But he can only use this argument by reasoning from a tangled perspective. II. First Person Singular and First Person Plural There are several different ways in which the first person pronoun may be used in philosophical prose. One such might be termed the authorial-executive, in which the writer declares what her intents and purposes are in the current writing. This remains a useful device for communicating strategy. At the moment, for example, I am exploring ways in which the first person can put in an appearance in philosophical texts. While this seems to me an interesting subject in its own right 24, I need not here attempt to be exhaustively systematic, since my main aim is to establish that Berkeley s use of the first person pronoun is not genuinely Cartesian (in a sense shortly to be explained).

12 12 Another way of using I might appropriately be labelled autobiographical. I may like potato chips, moonlight, and motor trips. But there is a live question as to whether you do too: your tastes might be very different. While common in other forms of speech and writing, ordinary autobiographical uses of the first person pronoun are rather rare in philosophical prose, presumably on the grounds of a shared consensus that they can contribute little or nothing in the way of argumentation. The avoidance of ordinary autobiographical usage conforms to disciplinary canons of relevance and helps maintain the austere purity of philosophical prose. The other use of I might be called exemplary: the writer takes her own case as an example which is used with implicit generality. This exemplary use is frequently encountered in philosophical prose. For example, we find Davidson writing: If I consider going to a certain concert, I know I will be put to a degree of trouble and expense, and I have more complicated beliefs about the enjoyment I will experience. 25 There is nothing to interest the biographer here. For the distinctive feature of the exemplary first person singular is its complete dispensability: the exemplary I could be replaced without significant change by we, or someone, or even any normal person. For this reason it is important not to confound an exemplary use with one in which the first person singular really does appear indispensably. And it is quite easy to be misled over the question whether the first person singular is being used indispensably or not. This is, for example, illustrated by a specious consideration sometimes advanced to support the view that folk psychology operates primarily through a process of simulation rather than by using a theory. It has often been urged in favour of the simulationist view that we at least know we sometimes use simulation to predict the behaviour of others, because we find ourselves asking questions like: What would I do in his situation?. However, it is far from clear this is the preamble to an episode of simulation, in the sense of trying out one s own responses with a view to projecting them onto others as predictions of behaviour. For such a question can equally be taken as a prompt for the application of general knowledge. We may be in the habit of saying or thinking to ourselves What would I do?, but this is an exemplary use which is no more than a stylistic variation. We could equally ask: What would any rational/normal person do in that situation?. So that isn t a good argument for simulationism, even though we surely do use simulation in trying to understand the minds of others. And Berkeley must have done so too, in imagining what ideas of sense other finite spirits have. 26 There is a further use of the first person singular which has a very special place in philosophical prose, and which is neither exemplary nor autobiographical. This is the Cartesian first person singular, as employed in a context of engagement with Cartesian scepticism. A philosophical writer undertaking to engage sceptical doubts such as those raised in Descartes s First Meditation may attempt to argue her way out of the epistemological plight produced by those doubts, as Descartes does, by arguing from claims based on what her ideas, thoughts, and experiences are like. It is an absolute rule of this philosophical game that one cannot fairly engage with scepticism on its own terms, without begging the very questions sceptical doubt raises, by moving without some reasoned justification from the first person singular to the first person plural.

13 13 Thus, it is not open to me to claim that I am not just dreaming that I am in a philosophy seminar because all of us here present are enjoying similar and interrelated experiences of a philosophical debate. Nor is it open to me to argue, in a Cartesian context, that this whiteboard and those marks I have made on this whiteboard are real because all of us in this room can see them, or at any rate have experiences as of a whiteboard with various marks on it. All that will be allowed me is that I have experiences as of a certain kind, and I will then have argumentative work to do in order to show there is something about these experiences which suffices to rule out such possibilities as that I am merely dreaming, or am being deceived by an Evil Demon, or am an envatted brain subject to direct synaptic stimulation. The experiences of other people, and the extent to which any such experiences are correlated with my own experiences, are things which need to be argued for on the basis of my own experiences. In short, the Cartesian context is one of solitary inquiry. Does Berkeley present his argument for God s existence in a Cartesian context? The Passivity Argument, as given above, could be advanced by a Cartesian reasoner. For it only makes claims about the reasoner s own ideas, observing that some of them are not dependent for their existence upon the reasoner s will, combining this with the general premise that ideas cannot exist without being dependent upon some mind or spirit. As already noted, this fails to satisfy the requirement that the ideas of sense produced in my mind should be the result of the activity of a single spirit with divine attributes. The argument can be strengthened by drawing attention to the coherence and orderliness of the ideas of sense in the reasoner s own mind. Let us allow that the degree of orderliness and coherence in these ideas is enough to justify an inference to their authorship by a single spirit far more powerful than the reasoner. It is important to appreciate that this major concession still fails to provide what Berkeley needs. Using this argument each reasoner may discover the best explanation of the orderliness and coherence in ideas of sense to be that they are produced by a single powerful spirit. So each may at least infer his own Demon. Atherton 27 draws attention to a passage in the Philosophical Commentaries, in which it may seem that Berkeley comes close to reasoning explicitly in this way: PC 838: Every sensation of mine which happens in consequence of the general, known Laws of nature and is from without i.e. independent of my Will demonstrates the Being of a God. i.e. of an unextended incorporeal Spirit wch is omniscient, omnipotent etc. So is this Berkeley s argument, an argument from the coherent and orderly nature of his own ideas of sense to their production by a spirit greater than himself? It might seem so, and this is the argument Atherton attributes to Berkeley, which I will label the Coherence of Sense Argument: The Coherence of Sense Argument 1. Ideas can only be caused by a spirit. 2. My ideas of sense are not caused by me. 3. There is a coherence and orderliness to my ideas of sense. Therefore, my ideas of sense are caused by God. 28

14 14 It might also be suggested that the attribution of the Shared Reality Argument to Berkeley is gratuitous since what he really offers us is a two-stage argument. The first stage is just the Passivity Argument: my ideas of sense are not caused by me, but must be due to spiritual agency. What s more, it s reasonable to conclude that in view of their orderly and coherent character they are produced by a single spirit. Isn t this the reasoning quite explicitly stated in a passage from the Second Dialogue in which Berkeley has Philonous distinguishing his views from those of Malebranche? This passage starts: Take here in brief my meaning. It is evident, that the things I perceive are my own ideas, and that no idea can exist unless it be in a mind. Nor is it less plain that these ideas, or things by me perceived, either themselves or their archetypes, exist independently of my mind, since I know myself not to be their author, it being out of my power to determine at pleasure, what particular ideas I shall be affected with upon opening my eyes or ears. They must therefore exist in some other mind, whose will it is that they should be exhibited to me. That completes stage one of the argument. After pausing to secure Hylas s assent to this conclusion, he moves on to stage two by drawing attention to the fact that while he can himself produce ideas these creatures of the fancy are not altogether so distinct, so strong, vivid, and permanent, as those perceived by my senses, which latter are called real things. From this he concludes that there is a mind which affects me every moment with all the sensible impressions I perceive. And furthermore from the variety, order and manner of these sensible impressions he infers the Author of them to be wise, powerful, and good, beyond comprehension. (Three Dialogues between Hylas and Philonous, Second Dialogue, The italics are Berkeley s.) So perhaps there is no justification for introducing the Shared Reality Argument, and Berkeley really does reason from his own ideas, in a way which conforms to the rules of Cartesian contexts; first to his ideas of sense being produced by some other spirit or spirits, and then to a single spiritual author as the best explanation of the order and coherence of his sensory experiences? No, that cannot give a satisfactory overall reading of Berkeley. At times, as in the above passage, this does appear to be the argument he explicitly advances. But we can see that he also has the Shared Reality Argument in mind. Our first reason for thinking this is that only the Shared Reality Argument can deliver the conclusion he seeks, the existence of God rather than the existence of a powerful spirit (who might be Descartes s Demon). But perhaps Berkeley did not appreciate this point and thought it sufficient to establish that his own ideas of sense had a single author? Yet even in the passages in which he starts in the first person singular he is liable to introduce some consideration which carries him outside a purely Cartesian context as in the reference to real things in the passage just quoted. For real things are not just Berkeley s own ideas of sense. Berkeley s dreams are ideas which he does not produce at will himself. But he would not argue from the occurrence of these ideas to the existence of God. In other words, Berkeley assumes distinctness, as Ayers noted: he holds that there are real objects which have an existence distinct from being perceived by me. Yet in a fully Cartesian context this could not be assumed, but is rather something which needs to be argued for.

15 15 And then there is the decisive consideration that Berkeley does not restrict himself to presenting the argument in the first person singular. On the contrary, he moves with apparent indifference between formulations in the first person singular and the first person plural, a practice frequently imitated by commentators and yet rarely mentioned by them. This was one of the textual features which at the outset we noted was something requiring explanation. Let us just swiftly run through the key passages in The Principles of Human Nature and the Three Dialogues between Hylas and Philonous to make the vacillating perspective apparent. The first occurrence of the argument for the existence of God is in paragraphs of The Principles of Human Nature. Paragraph 29 is in the singular:...i find the ideas actually perceived by sense have not a like dependence on my will.... There is therefore some other will or spirit that produces them. But immediately paragraph 30 transposes to the first person plural and the inference is clearly intended to be to the existence of a spirit who produces a shared reality: The ideas of sense are more strong, lively, and distinct than those of the imagination; they have likewise a steadiness, order, and coherence, and are not excited at random, as those which are the effects of human wills often are, but in a regular train or series, the admirable connexion whereof sufficiently testifies the wisdom and benevolence of its author. Now the set rules or established methods, wherein the mind we depend on excites in us the ideas of sense, are called the laws of nature: and these we learn by experience, which teaches us that such and such ideas are attended with such and such other ideas, in the ordinary course of things. Towards the conclusion of the Principles we again find the argument presented in the first person plural:... it is evident to every one, that those things which are called the works of nature, that is, the far greater part of the ideas or sensations perceived by us, are not produced by, or dependent on, the wills of men. There is therefore some other spirit that causes them, since it is repugnant that they should subsist by themselves. (PHK 146) In the Second Dialogue we find both singular and plural passages. The passage quoted above from pages is in the singular. But that follows hard on the heels of what Berkeley calls this easy reflexion : that the sensible world is that which we perceive by our several senses; and that nothing is perceived by the senses beside ideas; and that no idea or archetype of an idea can exist otherwise than in a mind. (Second Dialogue, 212-3) And in the Third Dialogue, in that solitary passage which might conceivably be taken as an expression of the Continuity Argument (pp.230-1), we find a mixture of singular and plural. Berkeley begins by claiming that since sensible things have an existence exterior to my mind there is some other mind wherein they exist, during the intervals between the times of my perceiving them. But he then goes on to conclude that since the same is true with regard to all other finite created spirits, it necessarily follows, there is an omnipresent, eternal Mind, which knows and comprehends all things, and exhibits them to our view in such a manner, and according to such rules as he himself hath ordained, and are by us termed the laws of nature. But of course in fact that does not follow a point we will shortly have to take up. It is clear from these shifts between singular and plural that Berkeley s use of the first person is not properly Cartesian. In using the first person singular in an exemplary way Berkeley is displaying quite a prominent feature of empiricist (Locke-Berkeley- Hume) philosophical reasoning. The objective is to give an account of human